Monday, May 31, 2004

A Remarkable Event in Justice IT history I have the honor of chairing an ad hoc group called the Global Training and Technology Assistance Committee (GTTAC) pulled together by the DOJ Office of Justice Programs and May 11-13 we sponsored one of the most significant events in the history of this field--a developer's workshop on the Global Justice XML Data Model (GJXDM).   Having a few weeks to ponder the significance of this event has just strengthened my conviction of how important this new model is but also how amazing this event was as a new paradigm of collaboration in the justice IT field.  A total of 325 developers and their managers attended this event, coming from 40 states and consisting of about an equal number of industry and government representatives.  The sight of an auditorium packed with people and each student desktop holding a laptop computer was awesome.  The computer scientists at the Georgia Tech Research Institute did all the heavy lifting in terms of providing spectacular instruction, but the endorsement of all the sponsoring groups and most notably the Office of Justice Programs was also impressive.   People were eager to learn about how to apply the model, chomping at the bit to go home with new knowledge to put it to work in computer-based information sharing, and more than willing to share their work with others.  What is also remarkable is that the whole thing was free, and that OJP did not award a specific grant to make this event happen.  The collaborating agencies figured out how to employ existing grants and other budgets to pay for some of the actual direct costs, and participants attended on their own nickel. I'm quite sure that nothing like this has ever happened before in this field.  There has never been a technical educational experience where developers who needed to know about a topic have been given such immersive training regardless of whether they labored for industry or government.  The sponsoring training and technical assistance organizations who came together to coordinate the event have never before been so like-minded about a need and so dedicated to its success.  All of this came to fruition within a few months after the initial conversations of the organizations that compose GTTAC.  The feedback on this training event has been extremely positive, although any such event can be improved.  What is most notable and should be the lesson learned for the future is that it is possible to build broad coalitions to deal with significant issues in criminal justice technology, and to do it quickly.   There will be a strong demand for follow-on events to this one, and its success and impact should be the premise for future partnerships in progress in this community.  8:41:01 PM    comment []

Technology pushed too far, too soon There is often the tendency in the early attempts to market an emerging tecnology to ask of it too much, and too soon.  Lessons abound in the too-early attempt to make pen-based computing the panacea for mobile computing, and other examples.  Worse than just disappointing early adoptors, the hype becomes so bad that the technology may never make it to market purely because of the negative perception of the general public.  This generally happens when technology is applied to something it is just not ready to do. One current technology suffering from early overexposure is facial recognition.  The populist excitement of having technology pick out a terrorist from the 100,000 people in a football stadium was so appealing and the entrepreneurs so eager for a big hit that they saw this chance at glory to powerful to resist.  Yet no real quantity of testing and preparation led anyone to believe that such fantastic capabilities were in the reach of the technology.  In this instance and in the trials where crowds walking down the street were scanned to find criminals, the technology was found wanting.  Because of the failures at these probably over-ambitious efforts, people get from the media that the technology does not work and therefore we should move on.  An example is the article by Barnaby J. Feeder in the New York Times entitled  Technology Strains to Find Menace in the Crowd. Despite growing acceptance, Feder contends there is "still a huge gap between the hype about the promise of face recognition technology and the results when it is put to the test."  [New York Times: Technology] The real lesson from this sitation is to use care to apply emerging technology to solve problems with its reach.   Facial recognition appears to be very successful in recognizing faces that were imaged during the booking process, where consistent lighting and grayscale standards were used.  Having this sort of standardized and properly lit set of images makes it possible to search the database of such images against suspects who may have just changed their name but have really been there before.  Success is also much greater in these instances even when using a lesser quality image as the search vehicle.  This technology has a tremendous potential to help law enforcement agencies match identities of suspects against previous databases, and its application should not be slowed by its failure to do what it was really not capabile of, at least for now. 8:07:04 PM    comment []

 <-- Return to Home Page

© Copyright 2005 Paul Wormeli. Last update: 3/16/2005; 9:24:41 PM.